[jgazal]

Well, it's at zero now, so a 100% increase would be fairly simple to accomplish. But do you have any idea what "typical consumer" means? There will never be that kind of technology in the home of a "typical consumer".  

The Comhear beamforming does not require customization, although I beleive it would perform better in rooms with controlled back reflections.

It actually seems like a soundbar.

You know what "every headphone listener" has access to now? Precision headphone-specific EQ developed from high resolution measurements. It's available on the most common headphone listening platform in the world, and is so cheap as to be a non issue.  Ever heard of that? Didn't think so.
(...)
I believe they are two different things and don't present the same way.
That's not a small problem, though, it's actually a deal-breaker.  And because it's hard to get it right without customizing, it severely limits the commercial application of the technology.  And yet, without a good listener HRTF in there somewhere, the whole thing just doesn't work very well.
I have not played much with Ambisonics, but from my binaural work, matching the HRTF to the listener is critical to the palpability of the resulting image. It's not so much a tonal issue as an imaging accuracy issue.  There's also an artistic problem too.  It isn't always desirable to have that full 3D sound field. 

Every time I listen through headphones to a Neumann KU-100 binaural recording the externalization is on the back of my head. Definitely my HRTF does not match the one from that dummy head microphone. But that is without any externalization DSP.

So my question is not about regular binaural.

It is about comparing Ambisonic content versus binaural encoded content with a Realiser (that uses a personal room impulse response set to not add crosstalk and headtracking for externalization).

Or if you prefer comparing binaural through loudspeakers (with "customized crosstalk cancellation - xtc" or a beamforming device without customization) versus third order ambisonics in regular listening rooms.

I do agree that custom xtc or 16 ambisonics speakers probably won't be mass marketed.

I also agree that custom headphone externalization device only could proliferate if it had easy HRTF acquisition.

But I am not so sure why you are so pessimist about beamforming devices that fire separate beams and therefore avoid crosstalk without customization.

Anyway, if the future scenario you described holds true, I would expect object based codecs (and maybe binaural encoded beds to reproduce the ambience) to become an standard, so that all those playbacks environments can interoperate with DSP.

Thank you very much for your elucidative post. Now I sort of understand why Dolby and DTS went the object based route.

 

[pinnahertz]

]

Every time I listen through headphones to a Neumann KU-100 binaural recording the externalization is on the back of my head. Definitely my HRTF does not match the one from that dummy head microphone. But that is without any externalization DSP.

So my question is not about regular binaural.

It is about comparing Ambisonic content versus binaural encoded content with a Realiser (that uses a personal room impulse response set to not add crosstalk and headtracking for externalization).

Or if you prefer comparing binaural through loudspeakers (with "customized crosstalk cancellation - xtc" or a beamforming device without customization) versus third order ambisonics in regular listening rooms.

I do agree that custom xtc or 16 ambisonics speakers probably won't be mass marketed.

I also agree that custom headphone externalization device only could proliferate if it had easy HRTF acquisition.

Ok. I no longer see a question, and you're starting to see the reality of mass marketing, so perhaps just move on?

But I am not so sure why you are so pessimist about beamforming devices that fire separate beams and therefore avoid crosstalk without customization.

Product cost vs percieved benefit. It's not a win until it's in the $400 sound bar. Ain't going to happen any time soon.

Anyway, if the future scenario you described holds true, I would expect object based codecs (and maybe binaural encoded beds to reproduce the ambience) to become an standard, so that all those playbacks environments can interoperate with DSP.

We have object oriented codes now, they aren't penetrating the consumer market well at all. Nothing "binaural" will ever achieve general market acceptance because of the commitment the listener has to make to realize the benefit, and the commitment the music producer must make to produce a double inventory of every record. Not to mention market confusion.

Thank you very much for your elucidative post. Now I sort of understand why Dolby and DTS went the object based route.

I can't understand how anything I've said would bring you to that understanding, but for better or worse, they've done it. But they are getting into very, very few homes, and frankly not the majority of theaters either. I'm not sure object-oriented is a flat out win in the market yet. I believe DTS is quite late to the party, though.  I'm sure many a producer has already said, "Why are we spending the time and budget on this when most people can't hear it?" ...and staying with 5.1.  If you recall, Dolby Stereo was just like that in the early 1980s, the tech was there, the benefit wasn't very visible, so lots of mono movies for quite a few years.

 

[jgazal]

We have object oriented codes now, they aren't penetrating the consumer market well at all. (...) because of the commitment the listener has to make to realize the benefit, and the commitment the music producer must make to produce a double inventory of every record.

(...)

I can't understand how anything I've said would bring you to that understanding, but for better or worse, they've done it. But they are getting into very, very few homes, and frankly not the majority of theaters either.

I am sorry to bother you once again.

I went to an atmos theater and it didn't seem a 3d sound field.

So I would never commit to ceiling speakers for just an immersive effect.

That said, the reason why the things you said led me to that conclusion is simple.

If mastering engineers don't want to provide binaural encoded material or b-format ambisonics and you have plenty of object-based content available, professors Edgard Choueiri (bacch), Peter Otto (Comhear) or Tony Hooley (dynasonix) just need to write a function to input the objects with the positions assigned by the mastering engineer and output a binaural encoded stream mixed with the untouched bed.

Perhaps playing Dolby Headphone or Dts headphone output through such devices would render satisfactory rendering of 3d sound field, if such codecs are based in a similar function.

Only time will tell and I must confess you may well be right.

Cheers.

 

[pinnahertz]

I am sorry to bother you once again.

I went to an atmos theater and it didn't seem a 3d sound field.

So I would never commit to ceiling speakers for just an immersive effect.

...unless you wanted to replicate what the creators heard in the final mix. But you should also understand, Atmos is an application of a technology. It could theoretically be adapted for other purposes, such as 3D sound. But it wasn't, and "immersive" and "3D" are two entirely different applications.

That said, the reason why the things you said led me to that conclusion is simple.

If mastering engineers don't want to provide binaural encoded material or b-format ambisonics

Hang on now, fair is fair. Mastering engineers don't make those decisions.

and you have plenty of object-based content available,

But you don't. Atmos has been used in film, there's little if any music only productions mixed in Atmos. So no, there really is no music content available in object-oriented formats.

professors Edgard Choueiri (bacch), Peter Otto (Comhear) or Tony Hooley (dynasonix) just need to write a function to input the objects with the positions assigned by the mastering engineer and output a binaural encoded stream mixed with the untouched bed.

Pretty sure it's not that simple, though.

Perhaps playing Dolby Headphone or Dts headphone output through such devices would render satisfactory rendering of 3d sound field, if such codecs are based in a similar function.

The Atmos/object oriented concept in film is intended to place objects in position amid an array of speakers placed mostly high, with a 5.1 bed. That's not really creating a specific 3D field, though. Given the purpose of Atmos and 5.1, Dolby Headphone can decode those formats into an acceptable headphone approximation of the original position of the objects and channels. Again, not necessarily a 3D sound field, though. That would be a different application and while possible, that's not the target of Atmos at this time.

Only time will tell and I must confess you may well be right.

Cheers.

Apart from my pessimism, and understanding of the audio market, I wish I were not right. I think excellent 3D sound with real, palpable imaging would be quite entertaining. It would be wonderful if some of the best recordings of all time could be presented that way. But then, from my first exposure to multichannel music in 1997, I wished and longed for 5.1 mixes of my favorites. Still wishing for most of them, and the ones I did get were often disappointing. But I'll go on record here as saying, with conviction, that two-channel stereo is totally inadequate, anything but "pure", and hobbled in many ways.  Remember that the original stereo experiments done by Bell Labs concluded that the minimum channel count for believable stereo was three.  We have two, and we are none the less stuck with it. 

 

[gregorio]

[1-A] I went to an atmos theater and it didn't seem a 3d sound field.

[1-B] So I would never commit to ceiling speakers for just an immersive effect.

[2] If mastering engineers don't want to provide binaural encoded material or b-format ambisonics and you have plenty of object-based content available ...

 
1. How do you arrive at conclusion B from statement A? How do you know the film you saw actually had an Atmos mix, maybe it was a standard 7.1 or 5.1 mix? If it was definitely an Atmos mix, maybe the film in question didn't use the format to it's full potential? A film really has to be shot to take full advantage of Atmos and that may not be appropriate for some films. Additionally, convention and the directors' wishes come into play; by convention relatively little is placed in the rear speakers, as it's usually considered undesirable to cause the audience to look behind and remove their eyes from the screen, the same philosophy could apply to ceiling speaker use. Some directors are very conservative in their use of the full soundfield, others are very adventurous. Atmos is still a relatively new format and directors are still learning it's capabilities and how to employ those capabilities to further their storytelling goals. In short, you appear to have fallen into the same trap as many audiophiles; confusing the container (format) with the contents of the container (the mix)!
 
2. Mastering engineers don't have plenty of object based content available, they don't have any at all in fact! If you're talking about Re-recording Mixers rather than Mastering Engineers, then in film we're talking about huge complex mixes, large and very expensive commercial dub stages specifically designed for theatrical mixes, which almost certainly don't have any headphone output capability and where the Mixers (engineers) have little/no experience of headphone mixing.
 
G
 
EDIT: I posted this before I was aware of pinnahertz's last response.

 

[jgazal]

(...)
Hang on now, fair is fair. Mastering engineers don't make those decisions.
(...) 

You are right.

Please excuse me.

I should have written "if content copyright owners decide to not distribute binaural encoded or b-format ambisonics content because they know 99% of their target audience don't own and won't buy compatible playback devices, then…".

Is similar to the dilemma of digital and analog FM broadcasting in radio stations.

1. How do you arrive at conclusion B from statement A? How do you know the film you saw actually had an Atmos mix, maybe it was a standard 7.1 or 5.1 mix? If it was definitely an Atmos mix, maybe the film in question didn't use the format to it's full potential? A film really has to be shot to take full advantage of Atmos and that may not be appropriate for some films. Additionally, convention and the directors' wishes come into play; by convention relatively little is placed in the rear speakers, as it's usually considered undesirable to cause the audience to look behind and remove their eyes from the screen, the same philosophy could apply to ceiling speaker use. Some directors are very conservative in their use of the full soundfield, others are very adventurous. Atmos is still a relatively new format and directors are still learning it's capabilities and how to employ those capabilities to further their storytelling goals. In short, you appear to have fallen into the same trap as many audiophiles; confusing the container (format) with the contents of the container (the mix)!

(...)

G

EDIT: I posted this before I was aware of pinnahertz's last response.

I see you answered before you were aware of pinnahertz response, but with all due respect to your reasoning, I feel like you were establishing an unfair hypothesis.

I should have elaborated the concept of 3d sound field or chosen a more precise terminology.

Imagine speakers in a dolby atmos arrangement. You can imagine that all the speakers are contained sit in the boundaries of an imaginary sphere. So to a 3d sound field shall reproduce what Dr. Choueiri describes as proximity:

(…) If, for instance, in the original soundfield a fly cicrles the head of the ideal listener during the recording, a listener of that recording played back through the two loudspeakers of a BACCH™ 3D Sound system will hear, simply and naturally, the same fly circling his or her own head. If, in contrast, the same recording is played through standard stereo or surround sound systems the fly will be perceived to be inside the loudspeakers or, through the artifice of the phantom image, in the limited vertical plane between the loudspeakers. (…)

https://www.princeton.edu/3D3A/PureStereo/Pure_Stereose2.html

In other words, the listener hears a virtual sound source as it was placed inside the boundaries fo that imaginary sphere and not beyond the speakers.

The theater I went is atmos enabled and there was a dolby atmos demo with raindrops and thunders being reproduced by the speakers. I believe dolby wants that demo to explore the full potential of the algorithm. I heard rain that I perceived as raindrops beyond the ceiling speakers.

I just cannot tell you for sure that the movie mix was originally an atmos mix or an upmix. But I am sure the ceiling speakers were playing loud and I didn't hear any sound that conveyed the concept of proximity.

That is my short experience.

Now I would like to ask you too, do you believe the very best atmos mixes can place virtual sources inside the boundaries of that imaginary sphere I have just described?

If the horizontal speakers are at the level of the listener's head which I define as zero degree elevation and ceiling speakers at positive 45 degrees of elevation, do you believe the very best atmos mixes can place virtual sources inside at negative elevations?

I agree with everything else you both said that I didn't quoted here. This is not a right and wrong challenge. At least to what concerns to me, it grew up in the last few of my posts as an attempt to understand what forces drive each player of the market.

Thank you both for sharing your valuable knowledge and experience!

 

[pinnahertz]

You are right.

Please excuse me.

I should have written "if content copyright owners decide to not distribute binaural encoded or b-format ambisonics content because they know 99% of their target audience don't own and won't buy compatible playback devices, then…".

No, it's actually far more complex than that. To distribute binaural, the material has to be recorded that way. And there starts a huge chain of decisions that the greater volume of recorded music just cannot comply with to begin with. The decision to even deal with binaural or it's relatives would completely drive the production process, and thus impact the creative process. It's not simple, and not going to happen.

Is similar to the dilemma of digital and analog FM broadcasting in radio stations.

How did you arrive at that one? The digital transmission path is the same as the analog path up to a certain point in the signal chain where they must be handled separately. They are not incompatible, do not require unique handing until the final audio processor and coding. Sorry, couldn't be more different than the binaural issue.

The theater I went is atmos enabled and there was a dolby atmos demo with raindrops and thunders being reproduced by the speakers. I believe dolby wants that demo to explore the full potential of the algorithm. I heard rain that I perceived as raindrops beyond the ceiling speakers.

You never...and I mean NEVER..hear raindrops overhead! Dolby has blown it with Atmos in many ways, the demo is but one.

I just cannot tell you for sure that the movie mix was originally an atmos mix or an upmix. But I am sure the ceiling speakers were playing loud and I didn't hear any sound that conveyed the concept of proximity.

Atmos mix. And yes, high speakers are pushed hard so it's obvious. It's a hyped marketing demo, and doesn't explore the full potential at all.  I have lots of issues with Atmos, but lets not do that here.

 

[catspaw]

I agree that accuracy is subjective when we put in a subjective listener (human).
Thats why you cant use accuracy "by ear", but need to use devices that wont change their mind based on the mood they are in or if they had their morning coffee or not.
 
The important point why sound science is often not important to the average joe buying a good headphone is because they dont know if they will like something that is shown in numbers (there is no direct correlation between graphs and "what I like").
 
So, its a bit more complex than that. Because most people wont have the money to have the testing gear required for good testing to compare it to what they hear, they decide that decisions made on subjective opinion is fine (in fact, many will defend that over objective data, mostly to not feel like they made "the cheap" decision... no one likes to admit they made a bad decision, even if consciously).
 
Because I find myself in this second group, I decided that the closest I can be to finding out if my HP is accurate is to use gaming as a benchmark for HPs.
We are obviously talking about Sound effects not soundtracks here.
 
Bad part is, I noticed that total realism is not what I want. (Well, I dont want to go deaf in 20 days from playing a shooter). So, I settled for something a bit more subjective, and so far I am quite ok with my "wrong decision".
 
I hope it stays that way.

 

[gregorio]

  [1] I agree that accuracy is subjective when we put in a subjective listener (human). Thats why you cant use accuracy "by ear", but need to use devices that wont change their mind based on the mood they are in or if they had their morning coffee or not.
 
[2] The important point why sound science is often not important to the average joe buying a good headphone is because they dont know if they will like something that is shown in numbers ...
 
[3] ... most people wont have the money to have the testing gear required for good testing to compare it to what they hear, they decide that decisions made on subjective opinion is fine (in fact, many will defend that over objective data, mostly to not feel like they made "the cheap" decision... no one likes to admit they made a bad decision, even if consciously).
 
[4] I decided that the closest I can be to finding out if my HP is accurate is to use gaming as a benchmark for HPs.
[4a] Bad part is, I noticed that total realism is not what I want. (Well, I dont want to go deaf in 20 days from playing a shooter). So, I settled for something a bit more subjective, and so far I am quite ok with my "wrong decision".

 
1. These two sentences contradict each other. Either you can have "accuracy by ear", in which case accuracy is subjective or you need to use "devices" to eliminate "the mood" and other biases, in which case accuracy is objective, not subjective.
 
2. That's patently not true. The average joe and most audiophiles WILL commonly identify SQ/accuracy with numbers. Higher numbers in terms of a higher price or in terms of say 24/192 vs 16/44.1 are just two of the most common/obvious examples.
 
3. Much of the testing gear required at that level costs very little. It's not really about the financial cost of testing equipment IMHO, it's about the cost in terms of time and effort, effectively they just can't be bothered and would rather just read a review or two. I've no problem with that approach, each to their own. I get annoyed though when they publicly advise others and argue that their approach is actually more accurate and science is wrong.
 
4. How does gaming provide a benchmark for accuracy? You are making a subjective opinion about your HP's reproduction of the subjective opinions of the sound effects designers, what has that got to do with accuracy?
4a. You seem to realise that you're talking about subjective opinion rather than realism/accuracy? No one is saying you've made a "wrong decision" or that buying what you subjectively like is in any way wrong. What I (and others) are saying would be "wrong" is that you couldn't truthfully state/advise that your headphones are accurate or that accuracy is subjective, all you could truthfully state is that you subjectively like your headphones (and have relatively little idea how accurate they actually are)!
 
G

 

[jgazal]

How did you arrive at that one? The digital transmission path is the same as the analog path up to a certain point in the signal chain where they must be handled separately. They are not incompatible, do not require unique handing until the final audio processor and coding. Sorry, couldn't be more different than the binaural issue.

I don't know where you live, but in my country there are fm stations with hybrid digital radio broadcasting and this is the dilemma I was referring to:

radioworld.com - look before you leap

radioworld.com - elevated hd power part II

Is it possible to broadcast Atmos through hybrid digital radio?

And binaural encoded content?

Is it possible to implement beamforming in car cabins?

You never...and I mean NEVER..hear raindrops overhead! Dolby has blown it with Atmos in many ways, the demo is but one.

(...)

It's a hyped marketing demo, and doesn't explore the full potential at all.

As you said, fair is fair.

Would you please explain the reasons that made you sure.

You could have said that you know what demo I was exposed to and that you have access to it and that you know that raindrops are assigned to play in wall/horizontal speakers and that the ceiling speakers plays sounds of different nature.

Or simple that you have experience with Atmos mixing and there is a recommendation or common practice to not assign raindrops to the ceiling speakers.

Is that the case? I am genuinely curious.

Please consider that I was not in my home, with the flawed acoustics of my living room and speakers placed in wrong spots to adapt to the furniture.

This was a professional home theater certified by Dolby to play Atmos. Still, I couldn't hear any sound that conveyed the concept of proximity, which was disappointing.

I am even more disappointed now thinking that raindrops were coming from the walls speakers and not overhead.

 

[pinnahertz]

I don't know where you live, but in my country there are fm stations with hybrid digital radio broadcasting and this is the dilemma I was referring to:

radioworld.com - look before you leap

radioworld.com - elevated hd power part II

I believe you've misinterpreted those articles. The actual FCC authorized power increase of the digital signal is 6dB not 10dB, and only if non-interference with other stations can be proven. There are quite a number of cities with high station count where that won't fly.
 
But it doesn't matter, this has absolutely nothing whatsoever in parallel with our discussion.

Is it possible to broadcast Atmos through hybrid digital radio?

No. The maximum HD Radio bitrate is 300kbps, Atmos works with a bed Dolby TrueHD 5.1 with a maximum of 18mbps.

And binaural encoded content?

Binaural has been broadcast over FM for decades. Any high quality two channel transmission or storage medium is capable of handling binaural material.  I'm not quite sure why this is being asked at this point.a

Is it possible to implement beamforming in car cabins?

The literal answer would be "yes", the practical answer would be dependant on the cost and effectiveness of the technology, probably "no".

As you said, fair is fair.

Would you please explain the reasons that made you sure.

Of what?

You could have said that you know what demo I was exposed to and that you have access to it and that you know that raindrops are assigned to play in wall/horizontal speakers and that the ceiling speakers plays sounds of different nature.

Exactly how many (rather expensive) Dolby Atmos home demo pieces do you think they've produced? Yes, I know what you heard, it's what everybody hears when they run the Dolby Atmos demo. Yes, it was mixed for Atmos. Yes, the raindrops were positioned overhead. My point was just that in their vigorous efforts to emphasis what Atmos can do, Dolby chose to place a natural sound in a location that is is never heard. The put raindrops overhead. You hear raindrops when they hit the ground unless you're in a forest (and though I don't recall the visual, I don't think that was the case.)  The demo is extremely hyped.  They want to make absolutely sure everyone can hear Atmos.

Or simple that you have experience with Atmos mixing and there is a recommendation or common practice to not assign raindrops to the ceiling speakers.
Is that the case? I am genuinely curious.

No.

Please consider that I was not in my home, with the flawed acoustics of my living room and speakers placed in wrong spots to adapt to the furniture.

This was a professional home theater certified by Dolby to play Atmos. Still, I couldn't hear any sound that conveyed the concept of proximity, which was disappointing.

The Atmos home theater implementation has some very serious issues that impact its performance. This isn't the place to elaborate. It does not mimic the theater version well. It isn't as "scalable" as Dolby would have us believe.

I am even more disappointed now thinking that raindrops were coming from the walls speakers and not overhead.

They weren't. You heard them from overhead. It was a bad choice for a hyped demo.

 

[catspaw]

   
1. These two sentences contradict each other. Either you can have "accuracy by ear", in which case accuracy is subjective or you need to use "devices" to eliminate "the mood" and other biases, in which case accuracy is objective, not subjective.
 
2. That's patently not true. The average joe and most audiophiles WILL commonly identify SQ/accuracy with numbers. Higher numbers in terms of a higher price or in terms of say 24/192 vs 16/44.1 are just two of the most common/obvious examples.
 
3. Much of the testing gear required at that level costs very little. It's not really about the financial cost of testing equipment IMHO, it's about the cost in terms of time and effort, effectively they just can't be bothered and would rather just read a review or two. I've no problem with that approach, each to their own. I get annoyed though when they publicly advise others and argue that their approach is actually more accurate and science is wrong.
 
4. How does gaming provide a benchmark for accuracy? You are making a subjective opinion about your HP's reproduction of the subjective opinions of the sound effects designers, what has that got to do with accuracy?
4a. You seem to realise that you're talking about subjective opinion rather than realism/accuracy? No one is saying you've made a "wrong decision" or that buying what you subjectively like is in any way wrong. What I (and others) are saying would be "wrong" is that you couldn't truthfully state/advise that your headphones are accurate or that accuracy is subjective, all you could truthfully state is that you subjectively like your headphones (and have relatively little idea how accurate they actually are)!
 
G

1. Thats my point (the contradiction).
2.I am talking about graphs and how they correlate to subjective opinion. Most people cant make the connection.
3.Its still additional costs. Not everyone has disposable income to spend on whatever they want.
4.Of course it is subjective, but I have to choose something as a testing method. 

 

[old tech]

   
So you're saying that a professional conductor doesn't know what instruments sound like?

Well that is a nice black or white deflection.  Perhaps your wife is just more accustomed to the sound of vinyl and that is her reality.
 
There is a reason why listeners of the classical genre in particular are more drawn to digital, it simply is more accurate with the correct instrument timbres, separation of instruments, realistic (ie not artificially wide) sound stage, and being able to capture the full dynamic range of symphonic orchestra.
 
I sit on a local symphony board and I know many conductors.  I have yet to come across one that would claim vinyl is better than CDs in reproducing instruments accurately. In fact, that is the main reason why CDs were invented and why they (and hi res) totally dominate the classical music market.  I myself have far from perfect pitch but I have yet to hear one record that accurately reproduces a long sustained piano note, as just one example.

 

[gregorio]

  1. Thats my point (the contradiction).
2.I am talking about graphs and how they correlate to subjective opinion. Most people cant make the connection.
3.Its still additional costs. Not everyone has disposable income to spend on whatever they want.
4.Of course it is subjective, but I have to choose something as a testing method. 

 
2. Graphs are just a way to display numbers and numbers do commonly correlate to subjective opinion for many/most people/audiophiles, as already mentioned. For example, we could make a graph showing a relationship between price and people's expectations of quality.
 
3. Not necessarily additional financial cost, there are various measurement tools available as freeware. There is a cost in terms of time and learning curve to use the measurement tools though and not many can be bothered to pay this price.
 
4. Indeed you do have to choose a method, as do we all. I've no objection to you choosing a purely subjective method but then: A. You can't make any claims about accuracy, only about what you like and one can't get around this fact by claiming that accuracy is subjective, B. You are therefore achieving the exact opposite of what you originally stated "I decided that the closest I can be to finding out if my HP is accurate ...", because the purely subjective method is likely to be pretty much the FURTHEST you "can be to finding out if your HP is accurate"!
 
G

 

[RRod]

  Well that is a nice black or white deflection.  Perhaps your wife is just more accustomed to the sound of vinyl and that is her reality.
 
There is a reason why listeners of the classical genre in particular are more drawn to digital, it simply is more accurate with the correct instrument timbres, separation of instruments, realistic (ie not artificially wide) sound stage, and being able to capture the full dynamic range of symphonic orchestra.
 
I sit on a local symphony board and I know many conductors.  I have yet to come across one that would claim vinyl is better than CDs in reproducing instruments accurately. In fact, that is the main reason why CDs were invented and why they (and hi res) totally dominate the classical music market.  I myself have far from perfect pitch but I have yet to hear one record that accurately reproduces a long sustained piano note, as just one example.

 
Re piano, wonder what you think of these tracks. There has been some bleed into classical from the latest vinyl craze, but it sure doesn't have anything to do with accurate sound reproduction.